--- title: I moved my AI coding agent into its own container, and... description: Running an AI coding agent directly on your personal machine means granting it broad access to your files, credentials, and shell. A practical alternative is... canonical: https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: I moved my AI coding agent into its own container, and now I don't care what it breaks | daily.dev og:description: Running an AI coding agent directly on your personal machine means granting it broad access to your files, credentials, and shell. A practical alternative is... og:url: https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f og:image: https://api.daily.dev/og/posts/PD40Zz89F.png og:image:alt: I moved my AI coding agent into its own container, and now I don't care what it breaks og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # I moved my AI coding agent into its own container, and now I don't care what it breaks **[XDA Developers](https://daily.dev/sources/xda-developers)** · 8 min read · 1 upvotes · 0 comments ## Summary Running an AI coding agent directly on your personal machine means granting it broad access to your files, credentials, and shell. A practical alternative is isolating the agent inside an unprivileged LXC container on a Proxmox server, where it can only touch what you explicitly put there, and the whole environment can be snapshotted and rolled back. The setup uses Hermes Agent with separate profiles, MiniMax M3 as the coding model, and Discord as the control interface — letting you send tasks from your phone and retrieve results later without ever opening a terminal. The container sits on the LAN with its own IP, so anything the agent builds and serves is immediately accessible on the local network. Models are swappable (GLM, DeepSeek) with minimal config changes, and the isolation means mistakes stay confined to a throwaway box. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I run an AI coding agent safely without risking my main machine? Run the agent inside an unprivileged LXC container on a server like Proxmox, giving it its own non-root user, limited file access, and capped CPU/memory/disk. Because the container sits on copy-on-write storage, you can snapshot it before letting the agent work and roll back if it breaks anything, confining any damage to a disposable container. _For developers isolating AI coding agents from their main system, daily.dev surfaces practical containerization approaches like this one._ ### What model is used to power the Hermes Agent coding profile in this container setup? MiniMax M3, a 428-billion-parameter frontier coding and agentic model, accessed through a MiniMax subscription rather than run locally. It handles lightweight jobs like scaffolding a Python script or summarizing a Git repo, and Hermes treats it as a swappable provider so it can be replaced with GLM 5.2 or DeepSeek V4 Pro with just a config change. _Developers comparing hosted coding models for lightweight agent tasks can track options like MiniMax M3 via daily.dev._ ### Why would binding a dev server to 0.0.0.0 inside a container be risky? Binding to 0.0.0.0 instead of localhost makes a dev server reachable from any device on the local network via the container's IP and port, not just from the machine running it. In this setup an AI agent scaffolded a Markdown editor and bound it to all interfaces, making it instantly accessible from a phone on the same LAN without any reverse proxy or port forwarding. _Developers debugging accidental network exposure from dev servers can find practical write-ups like this on daily.dev._ ## Similar posts on daily.dev - [My Coding Agents: Off My Laptop and Onto a Remote Server](https://daily.dev/posts/my-coding-agents-off-my-laptop-and-onto-a-remote-server-zxlgjxw5k) · Atomic Spin · 1 upvotes · 0 comments - [Docker Sandboxes: Run Claude Code and More Safely](https://daily.dev/posts/docker-sandboxes-run-claude-code-and-more-safely-6ydoupfk7) · Docker · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#proxmox](https://daily.dev/tags/proxmox) [View this post on daily.dev](https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"I moved my AI coding agent into its own container, and now I don't care what it breaks","url":"https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f"},"datePublished":"2026-07-09T14:35:27.859Z","dateModified":"2026-09-14T06:00:10.606Z","description":"Running an AI coding agent directly on your personal machine means granting it broad access to your files, credentials, and shell. A practical alternative is...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e1692ce3a25e0f9a206aa0b9e956f589?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e1692ce3a25e0f9a206aa0b9e956f589?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"XDA Developers","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"XDA Developers","logo":"https://media.daily.dev/image/upload/s--kCg6yyAP--/f_auto,q_auto/v1774964407/logos/xda-developers?_a=BAMAMiWQ0","url":"https://daily.dev/sources/xda-developers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,proxmox","timeRequired":"PT8M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"XDA Developers","item":"https://daily.dev/sources/xda-developers"},{"@type":"ListItem","position":3,"name":"I moved my AI coding agent into its own container, and now I don't care what it breaks"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/i-moved-my-ai-coding-agent-into-its-own-container-and-now-i-don-t-care-what-it-breaks-pd40zz89f#faq","mainEntity":[{"@type":"Question","name":"How can I run an AI coding agent safely without risking my main machine?","acceptedAnswer":{"@type":"Answer","text":"Run the agent inside an unprivileged LXC container on a server like Proxmox, giving it its own non-root user, limited file access, and capped CPU/memory/disk. Because the container sits on copy-on-write storage, you can snapshot it before letting the agent work and roll back if it breaks anything, confining any damage to a disposable container. For developers isolating AI coding agents from their main system, daily.dev surfaces practical containerization approaches like this one."}},{"@type":"Question","name":"What model is used to power the Hermes Agent coding profile in this container setup?","acceptedAnswer":{"@type":"Answer","text":"MiniMax M3, a 428-billion-parameter frontier coding and agentic model, accessed through a MiniMax subscription rather than run locally. It handles lightweight jobs like scaffolding a Python script or summarizing a Git repo, and Hermes treats it as a swappable provider so it can be replaced with GLM 5.2 or DeepSeek V4 Pro with just a config change. Developers comparing hosted coding models for lightweight agent tasks can track options like MiniMax M3 via daily.dev."}},{"@type":"Question","name":"Why would binding a dev server to 0.0.0.0 inside a container be risky?","acceptedAnswer":{"@type":"Answer","text":"Binding to 0.0.0.0 instead of localhost makes a dev server reachable from any device on the local network via the container's IP and port, not just from the machine running it. In this setup an AI agent scaffolded a Markdown editor and bound it to all interfaces, making it instantly accessible from a phone on the same LAN without any reverse proxy or port forwarding. Developers debugging accidental network exposure from dev servers can find practical write-ups like this on daily.dev."}}]} ```